git.maemo.org Git - jamaendo/blob - jamaendo/api.py

   1 import urllib, threading, os, gzip, time, json, re
   2 _DUMP_URL = '''http://img.jamendo.com/data/dbdump_artistalbumtrack.xml.gz'''
   3 _DUMP = os.path.expanduser('''~/.cache/jamaendo/dbdump.xml.gz''')
   4
   5 try:
   6     os.makedirs(os.path.dirname(_DUMP))
   7 except OSError:
   8     pass
   9
  10 def has_dump():
  11     return os.path.isfile(_DUMP)
  12
  13 def _file_is_old(fil, old_age):
  14     return os.path.getmtime(fil) < (time.time() - old_age)
  15
  16 def _dump_is_old():
  17     return not has_dump() or _file_is_old(_DUMP, 60*60*24) # 1 day
  18
  19 def refresh_dump(complete_callback, progress_callback=None, force=False):
  20     if force or _dump_is_old():
  21         downloader = Downloader(complete_callback, progress_callback)
  22         downloader.start()
  23     else:
  24         complete_callback()
  25
  26 class Downloader(threading.Thread):
  27     def __init__(self, complete_callback, progress_callback):
  28         threading.Thread.__init__(self)
  29         self.complete_callback = complete_callback
  30         self.progress_callback = progress_callback
  31
  32     def actual_callback(self, numblocks, blocksize, filesize):
  33         if self.progress_callback:
  34             try:
  35                 percent = min((numblocks*blocksize*100)/filesize, 100)
  36             except:
  37                 percent = 100
  38             self.progress_callback(percent)
  39
  40     def run(self):
  41         urllib.urlretrieve(_DUMP_URL, _DUMP, self.actual_callback)
  42         self.complete_callback()
  43
  44 def fast_iter(context, func):
  45     for event, elem in context:
  46         func(elem)
  47         elem.clear()
  48         while elem.getprevious() is not None:
  49             del elem.getparent()[0]
  50     del context
  51
  52 from lxml import etree
  53
  54 class Obj(object):
  55     def __repr__(self):
  56         def printable(v):
  57             if isinstance(v, basestring):
  58                 return v.encode('utf-8')
  59             else:
  60                 return str(v)
  61         return "{%s}" % (", ".join("%s=%s"%(k.encode('utf-8'), printable(v)) \
  62                              for k,v in self.__dict__.iteritems() if not k.startswith('_')))
  63
  64 class LocalDB(object):
  65     def __init__(self):
  66         self.fil = None
  67
  68     def connect(self):
  69         self.fil = gzip.open(_DUMP)
  70
  71     def close(self):
  72         self.fil.close()
  73
  74     def make_artist_obj(self, element):
  75         if element.text is not None and element.text != "":
  76             return element.text
  77         else:
  78             ret = {}
  79             for child in element:
  80                 if child.tag in ['name', 'id', 'image']:
  81                     ret[child.tag] = child.text
  82                 if child.tag == 'Albums':
  83                     albums = []
  84                     for album in child:
  85                         albie = {}
  86                         for albumitem in album:
  87                             if albumitem.tag in ['name', 'id']:
  88                                 albie[albumitem.tag] = albumitem.text
  89                         albums.append(albie)
  90                     ret['albums'] = albums
  91             return ret
  92
  93     def make_album_obj(self, element):
  94         if element.text is not None and element.text != "":
  95             return element.text
  96         else:
  97             ret = {}
  98             artist = element.getparent().getparent()
  99             if artist is not None:
 100                 for child in artist:
 101                     if child.tag == 'name':
 102                         ret['artist'] = child.text
 103                     elif child.tag == 'id':
 104                         ret['artist_id'] = child.text
 105             for child in element:
 106                 if child.tag in ['name', 'id', 'image']:
 107                     if child.text:
 108                         ret[child.tag] = child.text
 109                     else:
 110                         ret[child.tag] = ""
 111                 if child.tag == 'Tracks':
 112                     tracks = []
 113                     for track in child:
 114                         trackd = {}
 115                         for trackinfo in track:
 116                             if trackinfo.tag in ['name', 'id', 'numalbum']:
 117                                 trackd[trackinfo.tag] = trackinfo.text
 118                         tracks.append(trackd)
 119                     ret['tracks'] = tracks
 120             return ret
 121
 122     def artist_walker(self, name_match):
 123         for event, element in etree.iterparse(self.fil, tag="artist"):
 124             name = element.xpath('./name')[0].text.lower()
 125             if name and name.find(name_match) > -1:
 126                 yield self.make_artist_obj(element)
 127             element.clear()
 128             while element.getprevious() is not None:
 129                 del element.getparent()[0]
 130         raise StopIteration
 131
 132     def album_walker(self, name_match):
 133         for event, element in etree.iterparse(self.fil, tag="album"):
 134             name = element.xpath('./name')[0].text
 135             if name and name.lower().find(name_match) > -1:
 136                 yield self.make_album_obj(element)
 137             element.clear()
 138             while element.getprevious() is not None:
 139                 del element.getparent()[0]
 140         raise StopIteration
 141
 142     def artistid_walker(self, artistids):
 143         for event, element in etree.iterparse(self.fil, tag="artist"):
 144             _id = element.xpath('./id')[0].text
 145             if _id and int(_id) in artistids:
 146                 yield self.make_artist_obj(element)
 147             element.clear()
 148             while element.getprevious() is not None:
 149                 del element.getparent()[0]
 150         raise StopIteration
 151
 152     def albumid_walker(self, albumids):
 153         for event, element in etree.iterparse(self.fil, tag="album"):
 154             _id = element.xpath('./id')[0].text
 155             if _id and (int(_id) in albumids):
 156                 yield self.make_album_obj(element)
 157             element.clear()
 158             while element.getprevious() is not None:
 159                 del element.getparent()[0]
 160         raise StopIteration
 161
 162     def search_artists(self, substr):
 163         substr = substr.lower()
 164         return (artist for artist in self.artist_walker(substr))
 165
 166     def search_albums(self, substr):
 167         substr = substr.lower()
 168         return (album for album in self.album_walker(substr))
 169
 170     def get_artists(self, artistids):
 171         return (artist for artist in self.artistid_walker(artistids))
 172
 173     def get_albums(self, albumids):
 174         return (album for album in self.albumid_walker(albumids))
 175
 176 _GET2 = '''http://api.jamendo.com/get2/'''
 177
 178 class Query(object):
 179     last_query = time.time()
 180     caching = True
 181     cache_time = 60*60*24
 182     rate_limit = 1.0 # max queries per second
 183
 184     def __init__(self, order,
 185                  select=['id', 'name', 'image', 'artist_name'],
 186                  request='album',
 187                  track=['track_album', 'album_artist'],
 188                  count=5):
 189         if request == 'track':
 190             self.url = "%s%s/%s/json/%s?n=%s&order=%s" % (_GET2, '+'.join(select), request, '+'.join(track), count, order)
 191         else:
 192             self.url = "%s%s/%s/json/?n=%s&order=%s" % (_GET2, '+'.join(select), request, count, order)
 193
 194
 195     def _ratelimit(self):
 196         now = time.time()
 197         if now - self.last_query < self.rate_limit:
 198             time.sleep(self.rate_limit - (now - self.last_query))
 199         self.last_query = now
 200
 201     def __call__(self):
 202         """ratelimited query"""
 203         self._ratelimit()
 204         f = urllib.urlopen(self.url)
 205         ret = json.load(f)
 206         f.close()
 207         return ret
 208
 209     @staticmethod
 210     def album_cover(albumid, size=200):
 211         to = '~/.cache/jamaendo/cover-%d-%d.jpg'%(albumid, size)
 212         if not os.path.isfile(to):
 213             url = _GET2+'image/album/redirect/?id=%d&imagesize=%d'%(albumid, size)
 214             urllib.urlretrieve(url, to)
 215         return to
 216
 217     @staticmethod
 218     def track_ogg(trackid):
 219        return _GET2+ 'stream/track/redirect/?id=%d&streamencoding=ogg2'%(trackid)
 220
 221     @staticmethod
 222     def track_mp3(trackid):
 223        return _GET2+ 'stream/track/redirect/?id=%d&streamencoding=mp31'%(trackid)
 224
 225 class Queries(object):
 226     albums_this_week = Query(order='ratingweek_desc')
 227     albums_all_time = Query(order='ratingtotal_desc')
 228     albums_this_month = Query(order='ratingmonth_desc')
 229     albums_today = Query(order='ratingday_desc')
 230     playlists_all_time = Query(select=['id','name', 'user_idstr'], request='playlist', order='ratingtotal_desc')
 231     tracks_this_month = Query(select=['id', 'name',
 232                                       'stream',
 233                                       'album_name', 'artist_name',
 234                                       'album_id', 'artist_id'],
 235                               request='track',
 236                               order='ratingmonth_desc')